ASIAN J. EXP. BIOL. SCI.

VOL 2(1) 2011: 47-52

© Society of Applied Sciences

ORIGINAL ARTICLE

Prediction of Translycopene Binding Site of Lycopene Cleavage

Oxygenase Enzyme Involved in Bixin Synthetic Pathway: A
Computational Approach
a
Raghunath Satpathy , Rashmiranjan Behera, Rajesh ku.Guru and Aparajita Priyadarshini
Department of Biotechnology, MIRC LAB, MITS Engineering College, Rayagada, Odisha, India-765017.

ABSTRACT
Bixin is a natural dye and a high commercial important compound, produced from Bixin synthetic pathway in case of Bixa
orellana plant. The particular enzyme Lycopene cleavage Oxygenase catalyzes the first step of reaction pathways from Trans-
lycopene to Bixin synthesis. The 3D structure of the enzyme was predicted by MODELLER9v7 program. Model validation
was done by using the output of PROCHECK, DOPE score. The Ramachandran plot for the model was observed as 87.3
percentages of residues is in favourable regions that indicate the model is reliable. The docking study was performed with the
substrate Translycopene by AUTODOCK 4.0.2 tool. The binding residues were predicted by CAST P server as GLU 28, ASP
30, LYS 57, THR 59, SER 68, PHE 75, SER 76, GLN 81, PRO 82, and TYR 83. Further stability was realised by computing
energy value and RMSD (Root men square deviation) fluctuation of Carbon alpha back bone of the enzyme and the enzyme
with substrate complex, by performing molecular dynamics simulation using GROMACS tool.
KEYWORDS: Homology modelling, molecular dynamics, Bixin, model validation, docking, Substrate binding site

INTRODUCTION
Annatto (Bixa orellana L.) contains pigment contains bixin and norbixin, valuable natural colorants [1].These
pigments are widely used for industrial food and beverages, cosmetics, pharmaceutical products, and as natural dyes
for textiles [2, 3]. Restrictions are imposed on the use of synthetic additives in the food industry and investigation on
biochemical properties of annatto plants and seeds as it is having tumor inhibiting capacity [4].The natural dye bixin is
extracted from the seed of the annatto plant.Bixin is produced from the sub pathway of isoprenoid biosynthesis. More
specifically from the translycopene the bixin synthetic pathway diverges by the enzyme Lycopene cleavage
oxygenase and further the action of intermediate enzymes like Lycopene beta cyclase and Lycopene epsilon cyclase
lead to production of the pigment [5].So due to highly commercial and pharmaceutical importance of the compound
the cloning and protein engineering approach could be used to enhance the bixin production in industrial level.
The first enzyme i.e. Lycopene cleavage oxygenase that is involved in the bixin production pathway has been
considered for the present case. The present study aim to model and to find out the translycopene binding site of the 3D
structure of Lycopene cleavage oxygenase by computational approach whose structure has not been reported yet.
Then to check the reliability of the model and validation purposes various tools and molecular dynamics method is
used. The combined docking study and automated prediction server has been used to trace the substrate binding
pocket and the residues present in the translycopene binding site of the enzyme.

MATERIALSAND METHODS
Sequence retrieval and 3D model building
The synthetic pathway of bixin production was obtained from the Metacyc pathway database [6].The sequence for the

ASIAN J. EXP. BIOL. SCI. VOl 2 (1) 2011 47

Prediction of translycopene binding site ………………….......................................................................….. A computational approach R. Satpathy et al

Lycopene cleavage oxygenase enzyme was retrieved from Swissprot database from Expasy server [7] and a protein
BLAST search was performed against PDB (Protein Databank) to retrieve the corresponding template for the enzyme
[8]. The model was built by MODELLER9v7 program [9]. The MODELLER program uses an automated approach to
comparative protein structure modelling by satisfaction of spatial restraints. In brief the modelling procedure begins
with an alignment of the sequence to be modelled (target) with related known 3D structures (templates).The output is a
3D model for the target sequence containing all main chain and side chain non-hydrogen atoms [10]. The program also
employs probability density functions (PDFs) derived analytically using statistical mechanics and empirically using a
database of known protein structures as the spatial restraints.
Model validation
The MODELLER generated structure was analysed for any missing side chains during the process of model
construction by SCWRL4 tool [11] and also further verified by PROCHECK [12].The PROCHECK program
provides the information about the stereo chemical quality of a given protein structure and was used to generate
Ramachandrans plot. The DOPE (discrete optimized potential energy) score of modeller output per residues of the
model was observed. DOPE score is calculated by Modeller program which is the distance dependent statistical
potential based on probabilistic theory. This is extremely useful in making decisions about reliability [13].
Docking study and prediction of substrate binding site
In-silico docking was performed with the validated model and energy minimized translycopene structure [14]. For
that AUTODOCK 4.0.2 tool was used which is designed to predict how small molecules, such as substrates or drug
candidates, bind to a receptor of known 3D structure [15].To cross verify the binding site the model was also observed
in CASTp server [16] to figure out the substrate binding pocket regions of the enzyme. CASTp provides identification
and measurements of surface accessible pockets as well as interior inaccessible cavities, for proteins and other
molecules also the accessible surface area and volume of each pocket and cavity.
Molecular dynamics simulation by using High performance Computing facility
The stability of the model was realised by performing molecular dynamics simulation in by GROMACS 4.0.4
software in a high performance computing environment. The computing facility utilised is High performance cluster
for Biological Applications which is based on Intel Xeon Dual Quad core as processor, Gluster HPC 1.3 X86-64 bit
edition ,total 16 nodes each having 4GB of memory. For the calculation of RMSD (Root mean square deviation) and
Energy value of the model protein GROMACS 4.0.4 (Groningen Machine for Chemical Simulation) was used [17].
GROMACS is a versatile package to perform molecular dynamics, i.e. simulate the Newtonian equations of motion
for systems with hundreds to millions of particles [18]. Both the model and complex were subjected to molecular
dynamics simulation in water at 300 K temperature and for 2000 Pico second by using Gromos 43a1 force field of
GROMACS package.

RESULTSAND DISCUSSIONS
Homology modelling of Lycopene cleavage oxygenase and model validation
The considered suitable template for Lycopene cleavage oxygenase protein is 2BIW chain A [19] obtained from
BLAST search against PDB (Protein Data bank). Then Modeler9v7 program was used to generate the three
dimensional structure (Figure 1). Modelling was performed for 10 times and the output of the model having lowest
molpdf value was chosen as the final model. The model was observed that contains 369 amino acid residues with 5
helixes, 29 strands and 40 numbers of turns. To verify further the predicted structures, The DOPE score for the model
was obtained from Modeller output has been shown in Figure 2. From the figure the peak indicates that there is no
defect in the loop regions of the model. The validation of the model was carried out using Ramachandran plot
calculations computed with the PROCHECK program. The Phi and Psi distributions of the Ramachandran plot of
non-Glycine, non-Proline residues are summarized in Figure 3. Altogether 100% of the residues were in favoured and
allowed regions. The overall G-factor (Goodness factor) used was computed as -0.04 (Table 1). The Homology
modelling and validation results obtained were quite satisfactory.
Docking study and prediction of substrate binding site
Docking was performed between the Enzyme model as receptor and Translycopene (Ligand) by Autodock 4.0.2 Tool.
After Docking Was finished the result .dlg file was analysed to find the best ligand conformation. The docking energy
(binding energy +intermolecular energy) was calculated as -8.73. Also from the complex the binding site was
observed. Castp server was used to find the binding pocket of ligand and receptor. From the Castp server the binding

48 ASIAN J. EXP. BIOL. SCI. VOl 2 (1) 2011

Prediction of translycopene binding site ………………….......................................................................….. A computational approach R. Satpathy et al
site was predicted that contains residues like GLU 28, ASP 30, LYS 57, THR 59, SER 68, PHE 75, SER 76, GLN 81,
PRO 82, TYR 83, (Figure 4). The docking study of the substrate with the enzyme model show the binding residues
which are consistent with the result obtained from automated ligand binding pocket prediction tool CASTp.
Molecular dynamics simulation study
To check the stability of the enzyme model and enzyme with substrate after docking was subjected to molecular
dynamics simulation for 2000 Pico seconds which is a reliable method [20].The molecular dynamics simulation was
performed also both energy and root mean square deviation plots were derived from the respective trajectory file by
Gromacs software output. The total energy for the complex was observed within the range from -4.85e+05 to -
4.8e+05 KJ/mole and total energy for the enzyme model was from -4.75e+05 to -4.65e+05 KJ/mole (Figure 6). The
RMSD fluctuation plot shows the C-alpha backbone deviation during the simulation process is within the range 0.15-
0.34 nanometre (Figure 5). The result for root mean square deviation for the enzyme and complex indicates there is
very little deviation in the back bones during molecular dynamics simulation. Since it is a tolerable fluctuation in the
backbone hence it confirms the model protein is stable. Also total energy calculation indicates the complex shows less
negative value (more stable) in comparison to energy of model alone. So the combined observation of root mean
square deviation and energy provides strong evidence that the substrate translycopene has been lodged in its proper
binding pocket of the enzyme Lycopene cleavage oxygenase.

Table 1 Ramachandrans plot calculations on 3D model of Lycopene cleavage oxygenase protein computed with the
PROCHECK program

% residues in favourable regions 87.3

% residues in additional residue regions 9.7
% residues in generously regions 2.3
% residues in disallowed regions 0.6

% of Non Proline and non Glycine residues 100

Figure 1 The final considered model of Lycopene cleavage oxygenase from Modeller output

ASIAN J. EXP. BIOL. SCI. VOl 2 (1) 2011 49

Prediction of translycopene binding site ………………….......................................................................….. A computational approach R. Satpathy et al

Figure 2 Analysis of the DOPE score profile of the model obtained from Modeller program.

Figure 3 Ramachandran plot calculation of the psi/phi angle distribution of the model as computed by PROCHECK
program.

Figure 4 Predicted substrate binding pocket and residues of substrate translycopene (Green colour)

50 ASIAN J. EXP. BIOL. SCI. VOl 2 (1) 2011

Prediction of translycopene binding site ………………….......................................................................….. A computational approach R. Satpathy et al

Figure 5 Root mean square deviation plot (Carbon Alpha back bone) obtained from Gromacs tool during molecular
dynamics simulation for 2000 Pico second.

Figure 6 Energy plot obtained during 2000 Pico second molecular dynamics simulation of the model and complex at
300 K temperature by Gromacs tool.

CONCLUSION
The work basically describes a working algorithm to predict the substrate binding pockets and residues of an enzyme.
Here in this work, we propose a valid and stable 3D model of Lycopene cleavage oxygenase enzyme whose structure
is not present in the database.Positive results obtained from model validation process confirms our prediction.Also we
have predicted the key residues and substrate binding pocket of the substrate Translycopene. The docking result is
found to be consistent with the CastP server. The RMSD and energy calculation for the enzyme model and the
enzyme- substrate complex by molecular dynamics study shows the validity of the predicted binding site. However
this is a computational work further experimental analysis is needed and by using protein engineering approach it is
possible to enhance the production of the commercially important compound i.e. Bixin.

ASIAN J. EXP. BIOL. SCI. VOl 2 (1) 2011 51

Prediction of translycopene binding site ………………….......................................................................….. A computational approach R. Satpathy et al
ACKNOWLEDGEMENT
We are thankful to Chief executive and Dean of Majhighariani Institute of Technology & Science, Rayagada for
providing us MIRC lab for computing facility.

REFERENCES
[1]. Rodrigues SM, Soares VL, de Oliveira TM, GesteiraAS, Otoni WC & Costa MG (2007) Isolation and purification of RNAfrom t i s s u e s
rich in polyphenols, polysaccharides, and pigments of annatto (Bixa orellana L). Mol Biotechnology, 37(3):220-224.
[2]. Pritam Chattopadhyay1, Sandipan Chatterjee & Sukanta K. Sen (2008). Biotechnological potential of natural food grade biocolorants.
African Journal of Biotechnology. 7(17): 2972-2985.
[3]. Russell KR, Omoruyi FO, Pascoe KO & Morrison EY (2008). Hypoglycaemic activity of Bixa orellana extract in the dog. Methods Find
Exp Clin Pharmacol., 30(4):301-305.
[4]. Reddy MK, Alexander-Lindo RL & Nair MG (2005). Relative inhibition of lipid peroxidation, cyclooxygenase enzymes, and human
tumor cell proliferation by natural food colors. J Agric Food Chem., 53(23):9268-9273.
[5]. Florence Bouvier, Claude Suire, Jerome Muttererand & Camara (2003). Oxidative remodelling of chromoplast carotenoids: identification
of the carotenoid dioxygenase CsCCD and CsZCD genes involved in Crocus secondary metabolite biogenesis. Plant Cell. 15(1): 47-62.
[6]. Ron Caspi et.al (2010). The MetaCyc database of metabolic pathways and enzymes and the BioCyc collection of pathway/genome
databases. Nucleic Acids Research, 38: D473-D479.
[7]. Michel Schneider, Michael Tognolli & Amos Bairoch (2004). The Swiss-Prot protein knowledgebase and ExPASy: providing the plant
community with high quality proteomic data and tools. Plant Physiology and Biochemistry. 42:1013-1021.
[8]. Altschul SF, Gish W, Miller W, Myers EW & Lipman DJ (1990). Basic local alignment search tool. J. Mol. Biol., 215(3):403-410.
[9]. Eswar N, Eramian D, Webb B, Shen M & SaliA(2008). Protein structure modelling with MODELLER. Methods: Mol. Biol. 426:145-159.
[10]. SaliA& Blundell TL (1993). Comparative protein modelling by satisfaction of spatial restraints. Mol. Biol,.234 (3):779-815.
[11]. Georgii G, Maxim V. Shapovalov & Roland L. Dunbrack Jr. (2009). Improved prediction of protein side-chain conformations with
SCWRL4. Proteins Structure, Function and Bioinformatics, 77:778-795.
[12]. Laskowski RA, MacArthur MW, Moss DS & Thornton JM (1993). PROCHECK: a program to check the stereo chemical quality of protein
structures. J. Appl. Cryst., 26: 283-291.
[13]. Min-Yi Shen & Andrej Sali (2006). Statistical potential for assessment and prediction of Protein structures. Protein Science, 15:2507-
2524.
[14]. Schuettelkopf AW & vanAalten D M F (2004). PRODRG - a tool for high-throughput crystallography of protein-ligand complexes. Acta
Crystallographica., 60:1355-1363.
[15].Goodsell DS, Morris GM, Huey R & Olson AJ (1996). Automated docking of flexible ligands: applications of Auto Dock. J. Mol.
Recognit., 9(1):15.
[16]. Joe Dundas, Zheng Ouyang, Jeffery Tseng, Andrew Binkowski, Yaron Turpaz & Jie Liang. (2006). CASTp: computed atlas of surface
topography of proteins with structural and topographical mapping of functionally annotated residues. Nucl. Acids Res., 34:116-118.
[17].Vander spoel D, Erik Lindahl, Berk Hess, Gerrit Greoenhof,Alen E mark & Hermann JC (2005). GROMACS: Fast, Flexible and
Free. J.comp.chem., 26(16):1701-1718.
[18].Hess B, Kutzner C, Van Der Spoel D & Lindahl E (2008). GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable
Molecular Simulation. J Chem Theory Comput, 4(3):435-447.
[19].Kloer DP, Ruch S, Al-Babili S, Beyer P & Schulz GE (2005). The structure of a retinal-forming carotenoid oxygenase. Science, 308: 267-
269.
[20].Tatyana Mamonova,Brandon Hespenheide,Rachel Straub,M F Thorpe & Maria Kurnikova (2005). Protein flexibility using constraints
from molecular dynamics simulations. Phys. Biol., 2:137-147.

Corresponding Author: Raghunath Satpathy, Department of Biotechnology, MIRC LAB, MITS Engineering College,
Rayagada, Odisha, India-765017. E-mail: rnsatpathy@gmail.com

52 ASIAN J. EXP. BIOL. SCI. VOl 2 (1) 2011